1 / 25

Roundoff and Truncation Errors

Roundoff and Truncation Errors. Berlin Chen Department of Computer Science & Information Engineering National Taiwan Normal University. Reference: 1. Applied Numerical Methods with MATLAB for Engineers , Chapter 4 & Teaching material. Chapter Objectives (1/2).

lelandr
Download Presentation

Roundoff and Truncation Errors

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Roundoff and Truncation Errors Berlin Chen Department of Computer Science & Information Engineering National Taiwan Normal University Reference: 1. Applied Numerical Methods with MATLAB for Engineers, Chapter 4 & Teaching material

  2. Chapter Objectives (1/2) • Understanding the distinction between accuracy and precision • Learning how to quantify error • Learning how error estimates can be used to decide when to terminate an iterative calculation • Understanding how roundoff errors occur because digital computers have a limited ability to represent numbers • Understanding why floating-point numbers have limits on their range and precision

  3. Chapter Objectives (2/2) • Recognizing that truncation errors occur when exact mathematical formulations are represented by approximations • Knowing how to use the Taylor series to estimate truncation errors • Understanding how to write forward, backward, and centered finite-difference approximations of the first and second derivatives • Recognizing that efforts to minimize truncation errors can sometimes increase roundoff errors

  4. Accuracy and Precision • Accuracy refers to how closely a computed or measured value agrees with the true value, while precision refers to how closely individual computed or measured values agree with each other • inaccurate and imprecise • accurate and imprecise • inaccurate and precise • accurate and precise

  5. Error Definitions (1/2) • True error (Et): the difference between the true value and the approximation • Absolute error (|Et|): the absolute difference between the true value and the approximation • True fractional relative error: the true error divided by the true value • Relative error (t): the true fractional relative error expressed as a percentage 10 → 9 10,000 → 9,999

  6. Error Definitions (2/2) • The previous definitions of error relied on knowing a true value. If that is not the case, approximations can be made to the error • The approximate percent relative error can be given as the approximate error divided by the approximation, expressed as a percentage - though this presents the challenge of finding the approximate error! • For iterative processes, the error can be approximated as the difference in values between successive iterations

  7. Using Error Estimates • Often, when performing calculations, we may not be concerned with the sign of the error but are interested in whether the absolute value of the percent relative error is lower than a prespecified tolerance s • For such cases, the computation is repeated until | a |< s • This relationship is referred to as a stopping criterion • Note that for the remainder of our discussions, we almost always employ absolute values when using relative errors • We say that an approximation is correct to at least n significant figures (significant digits) if its | a | is smaller than s that has a value

  8. Example: Exponential Function • It is known that the exponential function can be computed using • Try to add terms (1, 2, …, n) until the absolute value of the approximate error estimate | a | falls below a prescribed error criterion s conforming to three significant figures

  9. Roundoff Errors • Roundoff errorsarise because digital computers cannot represent some quantities exactly. There are two major facets of roundoff errors involved in numerical calculations: • Digital computers have size and precision limits on their ability to represent numbers • Certain numerical manipulations are highly sensitive to roundoff errors

  10. Example: a 10-based Floating-point System • If 0.03125 is represented by the system as 3.1X10-2 , a roundoff error is introduced • The roundoff error of a number will be proportional to its magnitude magnitude of the exponent Significand (mantissa)

  11. Computer Number Representation • By default, MATLAB has adopted the IEEE double-precision format in which eight bytes (64 bits) are used to represent floating-point numbers: n = ±(1+f) x 2e • The sign is determined by a sign bit • The mantissaf is determined by a 52-bit binary number • The exponent e is determined by an 11-bit binary number, from which 1023 is subtracted to get e Binary numbers consist exclusively of 0s and 1s. When normalized, the leading bit (always 1) does not have to be stored. Only the fractional part of the significand is stored.

  12. Floating Point Ranges • Values of -1023 and +1024 for e are reserved for special meanings, so the exponent range is -1022 to 1023 • The largest possible number MATLAB can store has • f of all 1’s, giving a significand of 2 - 2-52, or approximately 2 • e of 111111111102, giving an exponent of 2046 - 1023 = 1023 • This yields approximately 21024 ≈ 1.799710308 • The smallest possible number MATLAB can store with full precision has • f of all 0’s, giving a significand of 1 • e of 000000000012, giving an exponent of 1-1023 = -1022 • This yields 2-1022 ≈ 2.225110-308

  13. Floating Point Precision • The 52 bits for the mantissa f correspond to about 15 to 16 base-10 digits. The machine epsilon () - the maximum relative error between a number and MATLAB’s representation of that number, is thus 2-52 = 2.220410-16

  14. Roundoff Errors with Arithmetic Manipulations • Roundoff error can happen in several circumstances other than just storing numbers - for example: • Large computations- if a process performs a large number of computations, roundoff errors may build up to become significant • Adding a Large and a Small Number- Since the small number’s mantissa is shifted to the right to be the same scale as the large number, digits are lost • Smearing - Smearing occurs whenever the individual terms in a summation are larger than the summation itself • (x + 10-20) - x = 10-20 mathematically, but x = 1; (x + 10-20) - x gives a 0 in MATLAB!

  15. Truncation Errors • Truncation errorsare those that result from using an approximation in place of an exact mathematical procedure • Example 1: approximation to a derivative using a finite-difference equation: • Example 2: The Taylor Series

  16. The Taylor Theorem and Series • The Taylor theoremstates that any smooth function can be approximated as a polynomial • The Taylor series provides a means to express this idea mathematically • A good problem context is to use Taylor series to predict a function value at one point in terms of the function value and its derivatives at another point

  17. The Taylor Series

  18. The Taylor Series: Remainder Term

  19. More on Truncation Errors • In general, the nth order Taylor series expansion will be exact for an nth order polynomial • Any smooth function can be approximated as a polynomial • In other cases, the remainder term Rn is of the order of hn+1, meaning: • The more terms are used, the smaller the error, and • The smaller the spacing, the smaller the error for a given number of terms

  20. Numerical Differentiation (1/2) • The first order Taylor series can be used to calculate approximations to derivatives: • Given: • Then: • This is termed a “forward” difference because it utilizes data at i and i+1 to estimate the derivative

  21. Numerical Differentiation (2/2) • There are alsobackwardand centered difference approximations, depending on the points used: • Forward: • Backward: • Centered

  22. Total Numerical Error • The total numerical erroris the summation of the truncation and roundoff errors • The truncation error generally increasesas the step size increases, while the roundoff error decreases as the step size increases - this leads to a point of diminishing returns for step size

  23. Finite-Difference Approximation of Derivatives • Given a function: • we can use a centered difference approximation to estimate the first derivative of the above function at x=0.5 • However, if we progressively divide the step size by a factor of 10, roundoff errors become dominant as the step size is reduced

  24. Other Errors (1/2) • Blunders - errors caused by malfunctions of the computer or human imperfection • In early years of computers, erroneous numerical results could sometimes be attributed to malfunctions of the computer itself • Today, most blunders must be attributed to human imperfection • Can be avoided only by sound knowledge of fundamental principles and by the care when approaching and designing our solutions to the problem • Model errors - errors resulting from incomplete mathematical models • When some latent effects are not taken into account or ignored

  25. Other Errors (2/2) • Data uncertainty - errors resulting from the accuracy and/or precision of the data • When with biased (underestimation/overestimation) or imprecise instruments • We can use descriptive statistics (viz. mean and variance) to provide a measure of the bias and imprecision • For most of this course, we will assume that we have not made gross errors (blunders), we have a sound model, and we are dealing with error-free measurements

More Related